Skip to content

Fix CUDA version handling in ci_install_deepep.sh #14854

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merrymercy merged 2 commits into from
Dec 11, 2025
Merged

Fix CUDA version handling in ci_install_deepep.sh #14854

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e49d179
Fix CUDA version handling in ci_install_deepep.sh
merrymercy Dec 11, 2025
95745b1
Fix comment on CUDA version handling in script
merrymercy Dec 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion scripts/ci/ci_install_deepep.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,13 @@ if [ "$GRACE_BLACKWELL" = "1" ]; then
if [ "$CUDA_VERSION" = "12.8" ]; then
CHOSEN_TORCH_CUDA_ARCH_LIST='10.0'
elif awk -v ver="$CUDA_VERSION" 'BEGIN {exit !(ver > 12.8)}'; then
CHOSEN_TORCH_CUDA_ARCH_LIST='10.0;10.3'
# With cuda > 12.8, the compiler supports 10.3, so we should use
# CHOSEN_TORCH_CUDA_ARCH_LIST='10.0;10.3'
#
# However, our CI machine has a weird setup and nvidia-smi reports wrong CUDA version in the container.
# The container is actually cuda 12.8, but nvidia-smi reports 13.0, leading to compilation errors. so we
# drop 10.3.
CHOSEN_TORCH_CUDA_ARCH_LIST='10.0'
else
echo "Unsupported CUDA version for Grace Blackwell: $CUDA_VERSION" && exit 1
fi && \
Expand Down
Loading